Core KPIs
Default values are illustrative — adjust to match your game
Average ad impressions per user per day
% of DAU who watch any ad
More sessions = eCPM depreciation resets each session
Maximum ad positions per session — controls funnel and depreciation curve length
Market Mix
Example values — enter your own data
Retention
Used for LTV projection — enter your game's retention
Profitability
Enter CPI to reveal a pROAS breakeven chart
Cost per install — leave blank to hide breakeven overlay
Cumulative ad LTV ÷ CPI target (e.g. 150% = 1.5× CPI returned)
Daily Ad Revenue
$394.48
Monthly
$11,834.39
Daily Impressions
26,577
Ad Watchers
12,053
ARPDAU
$0.02
Effective eCPM
$14.84
Revenue by Impression Position
All daily ad positions shown — eCPM depreciation resets at the start of each session
| Position | Ad Watchers | Imps | eCPM | Revenue | Cumulative |
|---|---|---|---|---|---|
| Ad #1 | 12,053 | 12,053 | $16.00 | $192.85 | $192.85 |
| Ad #2 | 11,450 | 11,450 | $14.10 | $161.40 | $354.25 |
| Ad #3 | 3,074 | 3,074 | $13.09 | $40.23 | $394.48 |
| Total | — | 26,577 | $14.84 | $394.48 | $394.48 |
Ad LTV by IMPDAU
Cumulative ad revenue per install — fewer ads retain users longer
Fewer ads
$0.12
30-day Ad LTV / install
Current
$0.21
30-day Ad LTV / install
More ads
$0.28
30-day Ad LTV / install
eCPM Depreciation & Revenue by Position
How eCPM and revenue change with each additional ad shown per session
Ad Frequency Sweet Spot
Most revenue comes from the first few impressions — more ads past the sweet spot hurt retention without meaningful revenue gain
Metric Data
Using sample data (30 days)Upload or paste a row-level ad-revenue report to diagnose your own data. The diagnosis below runs on the sample series until you do.
Metric Timeline
Click a point to set the before/after split — or overlay dated events to associate a move with a cause
Event Timeline
No events yet. Add a campaign, SDK change, or release above — or import a CSV (columns: date, label, category, note).
Modeled LTV Impact
ModeledValue a before/after move on one driver as a modeled LTV change per install, over a 90-day horizon.
Held fixed: eCPM, view rate, sessions/day, base retention curve.
Retention: Retention adjusted by the engine's ad-load penalty (each impression/DEU above baseline reduces retention multiplicatively); base D1/D7/D14/D30 held fixed.
Modeled under stated assumptions — not a measured causal effect.
This is a modeled estimate. A randomized holdout or geo test-control experiment would convert it into a measured causal effect.
Ad Monetization Diagnosis
ARPDAU fell −5.8% · 2025-09-01 → 2025-09-30
$0.1025 → $0.0965 · move detected at the largest single-day change
What changed, and what drove it
| Driver | Contribution to ΔARPDAU | Share |
|---|---|---|
| Impressions / user | $0.0000 | 0% |
| eCPM (price)dominant | −$0.0060 | 100% |
| Interaction | $0.0000 | 0% |
| Total move | −$0.0060 | 100% |
What was happening around the move
No events are annotated in this window. Add campaign, SDK/mediation, release, or experiment markers on the timeline to attribute the move to a cause.
What the move is worth, per install
$0.1876 → $0.1766 per install · −5.9% over 90 days
- Held fixed:
- IMPDAU, view rate, sessions/day, retention curve.
- Retention:
- Retention held fixed — eCPM does not affect retention in this model.
Modeled under stated assumptions — not a measured causal effect.
Ranked next actions
- 1Blended eCPM moved −5.8% ($11.14 → $10.49) and is the dominant driver. Audit the mediation waterfall and floor prices — an eCPM drop of this size usually traces to a network/demand or floor change.
- 2Modeled LTV impact: −$0.01 per install over 90 days (−5.9%). Modeled under stated assumptions — not a measured causal effect.
- 3To move from a modeled estimate to a measured causal effect, run a randomized holdout / geo test-control on the change.
Want a deeper analysis for your game?
I help studios model ad strategy, forecast LTV, and find the revenue levers that actually move the needle.
Let's connect on LinkedIn →